rhamnolipid foam enhanced remediation of cadmium

RHAMNOLIPID FOAM ENHANCED REMEDIATION OF CADMIUMAND NICKEL CONTAMINATED SOIL

SUILING WANG and CATHERINE N. MULLIGAN∗Department of Building, Civil and Environmental Engineering, Concordia University, 1455 de

Maisonneuve Boulevard W., ER 303, Montreal, Quebec, H3G 1M8, Canada(∗author for correspondence; e-mail: [email protected], Tel: +1-514-848-2424,

Fax: +1-514-848-2809)

(Received 27 October 2003; accepted 24 March 2004)

Abstract. Column experiments were conducted to evaluate the feasibility of using a rhamnolipidfoam to remove heavy metals (Cd and Ni) from a sandy soil contaminated with Cd (1706 ppm) and Ni(2010 ppm). Best results were obtained from the foam generated by a 0.5% rhamnolipid solution withan initial pH value of 10.0 after flushing with 20-pore-volume of solution. These conditions removed73.2% of the Cd and 68.1% of the Ni. Removal efficiencies by foam generated by a chemical surfactant,Triton X-100, were investigated as a comparison. It removed 65.5% of the Cd and 57.3% of the Niunder the same conditions. After a 20-pore-volume liquid solution flushing, 0.5% rhamnolipid (initialpH 10.0) without foam generation removed 61.7% of the Cd and 51.0% of the Ni, whereas 0.5%Triton X-100 (initial pH 10.0) removed 52.8% of the Cd and 45.2% of the Ni. Distilled water withadjusted pH only was also used to flush through the contaminated soil column as a control. It removed17.8% of the Cd and 18.7% of the Ni. This study shows that rhamnolipid foam technology can be aneffective means for the remediation of cadmium and nickel contaminated soil.

Keywords: biosurfactant, cadmium, foam, nickel, remediation, rhamnolipid, soil, surfactant, TritonX100

1. Introduction

Heavy metals pose a persistent problem at many contaminated sites and are beingadded to soil, water, and air in increasing amounts from a variety of sources, in-cluding industrial, agricultural and military activities, and domestic effluents. Asa result, they are now widely dispersed in the environment in a range of differ-ent physicochemical forms (Roundhill, 2001), and are a source of some concernbecause of their potential reactivity, toxicity, and mobility in the soil (Selim andAmacher, 1996). Thus, heavy metals are included on the EPA’s list of prioritypollutants (Mulligan et al., 2001a).

A number of remediation technologies have been developed for heavy metalcontaminated soils such as soil excavation, thermal extraction for volatile metals(e.g. mercury, arsenic and cadmium as well as their compounds can be evapo-rated at 800 ◦C), electrokinetics, solidification/stabilization, vitrification, chemicaloxidation, soil flushing, and bioremediation (Mulligan et al., 2001b). The spe-cific technology selected for the treatment of a contaminated site depends onthe speciation of the contaminants and other site-specific characteristics. Another

Water, Air, and Soil Pollution 157: 315–330, 2004.C© 2004 Kluwer Academic Publishers. Printed in the Netherlands.

316 S. WANG AND C. N. MULLIGAN

important consideration is that the chosen method does not leave toxic residues,which must be subsequently removed (Roundhill, 2001). One or more of these ap-proaches are often combined together for more efficient and cost-effective treatment(Evanko and Dzombak, 1997).

Bioremediation has potential for the remediation of heavy metal contaminatedsites. It has been demonstrated that biosurfactants generated by bacteria and yeastscould be used potentially for the environmental remediation of heavy metals fromsoils and sediments (Miller, 1995; Torrens et al., 1998; Mulligan et al., 1999, 2001a).Biosurfactants can enhance the mobility of heavy metals by reducing the interfacialtension between the metals and soil and by forming micelles. One more attractivecharacteristic is that they are natural products, may be less toxic to biodegradingbacteria and can be degradable themselves. Therefore, they present effective andnontoxic candidates for the remediation of the contaminated sites. However, theuse of biosurfactant solutions has been limited by a number of parameters, suchas channeling effects, aqueous-phase bypassing, and rate limiting mass transfer(Rothmel et al., 1998). Moreover, it is difficult to control the migration of the fluidscontaining dissolved contaminants (Vignon and Rubin, 1989), which probably leadsto the spreading of the contaminated zone.

Biosurfactant foam is being developed as a promising substitute. It not onlyinherits all the merits and remedies the deficiencies, but also improves the removalefficiency. Foams display properties that are vastly different from the liquids, whichconstitute the foam (Chowdiah et al., 1998). The simultaneous injection of surfac-tant and air will enhance the flooding efficiency of surfactant flushing even in aheterogeneous porous medium, resulting in higher removal efficiency (Jeong et al.,2000). The use of foam can also provide a better control on the volume of flu-ids injected and the ability to contain the migration of contaminant-laden liquids(Chowdiah et al., 1998). At the same time, it could be more cost effective due tothe low usage of chemicals and surfactants. Surfactant foam technology has beeninvestigated to remove hydrophobic organic compounds, such as polynuclear aro-matic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), pentachlorophenol(PCP), and other chlorinated hydrocarbons from contaminated soils (Peters et al.,1994; Kilbane et al., 1997; Rothmel et al., 1998; Jeong et al., 2000; Mulligan andEftekhari, 2003).

The biosurfactant that was used in this study, a rhamnolipid, was from theglycolipid group and was produced by Pseudomonas aeruginosa (Tsujii, 1998).There are four types of rhamnolipids (Tsujii, 1998). Type 1 (R1) is L-rhamnosyl-β-hydroxydecanoyl-β-hydroxydecanoate of molecular mass 504 Da. Type II (R2) isL-rhamnosyl-β-L-rhamnosyl-β-hydroxydecanoyl-β-hydroxydecanoate of molec-ular mass 650 Da. The other two types of rhamnolipids contain either two rhamnosesattached to β-hydroxydecanoic acid, or one rhamnose connected to the identicalfatty acid. Rhamnolipid type I and type II are suitable for soil washing and heavymetal removal. While type III is for metal processing, leather processing, lubricants,pulp and paper processing, type IV is usually used in textiles, cleaners, foods, inks,

FOAM ENHANCED REMEDIATION OF CONTAMINATED SOIL 317

paints, adhesives, personal care products, agricultural adjuvants, and water treat-ment (JENEIL 2001).

In this study, experiments were first done to test the foamability and foam stabil-ity of JBR425 (mixed rhamnolipids) solutions. Next, the pressure gradient build-upin the soil column during foam passing through the column was investigated. Then,experiments were conducted to evaluate the feasibility of rhamnolipid foam en-hanced remediation of the heavy metal contaminated soil. Comparisons of themetal removal abilities were made between the biosurfactant, JBR425, and thechemical surfactant, Triton X-100.

2. Materials and Methods

2.1. BIOSURFACTANT/SURFACTANT

The biosurfactant, JBR425 (mixed rhamnolipids), was obtained from JENEILBiosurfactant Co. (USA). Rhamnolipids are more favored because of their lowtoxicity, high biodegradability, and low surface tensions. The rhamnolipids, usedin this study, were biosurfactants from the glycolipid group made by Pseudomonasaeruginosa with the trademark JBR425 from JENEIL Biosurfactant Co. JBR425is an aqueous solution of rhamnolipid at 25% concentration. It is produced froma sterilized and centrifuged fermentation broth. Two major types of rhamnolipids,RLL (R1) and RRLL (R2), are present in the solution. Several tests done by themanufacturer and independent laboratories (OECD 209ASRIT, OECD 301D andOECD 202) show the degree of biodegradability and toxicity of JBR215 meet theEPA requirements (JENEIL, 2001). The synthetic surfactant, Triton X-100, wasobtained from Sigma Chemical Co. (U.S.A.). The Triton X-series of nonionic sur-factants is prepared by the reaction of octylphenol with ethylene oxide. Both ofthe two surfactants are commercially available and are widely used as householdand industrial detergents and in other applications. They have also been studied inenvironmental applications (Mulligan et al., 1999, 2001a; Mulligan and Eftekhari,2003). The characteristics of these surfactants obtained from the material safetydata sheet (MSDS) and other literature are described in Table I.

2.2. SOIL TREATMENT

The soil sample was obtained from a building site in Montreal, Canada. It waswashed with distilled water and then dried in an oven at 105 ◦C for 48 h. Large par-ticles were crushed by mortar and pestle, and then mixed with fine silica sand in theratio of 1:4. Consequently, sieve analyses were done by using a set of USA standardtesting sieves. The particles that passed through the No. 200 sieve were discarded be-cause they are not applicable for the column tests. Then the soil was contaminatedartificially in the laboratory with metal salts, in the forms of NiCl2 · 6H2O andCd(NO3)2 · 6H2O, which were dissolved with distilled water (4000 mg/L of Cd and


TABLE I

Characteristics of surfactants used in the experiments

Surface tensionSurfactant Type Chemical formula M.W. (g) CMC (g/l) (mN/m)

JBR 425 Anionic C26H48O9, C32H58O13 573 0.03a 26a

Triton X-100 Nonionic C8H17C6(OC2H4)nOH 620 0.4 33(n = 9–10)

JENEIL (2001) and Sigma (1993).aMulligan et al. (2001a).

TABLE II

Characteristics of the contaminated soil

Organic matter content (%) 1.0%

Cation exchange capacitya (cmoles+/kg) 7.9 (pH = 7.0)

Hydraulic conductivity (cm/s) 0.02

Heavy metal content (mg metal/kg dry soil) Cd, Ni 1706, 2010

Sieve analysis results 84.7% sand and 15.3% silt

Specific gravity 2.43

aGillman and Sumpter (1986).

4000 mg/L of Ni) and were added together into the soil sample (without adjustingthe pH value) at the same time. The soil was left in the solution over two weeks, andthen the suspension was shaken on a wrist action shaker at 60 oscillations/min for24 h. at room temperature (25.0 ± 0.2 ◦C), and subsequently centrifuged. The super-natant was discarded and the contaminated soil was dried at 105 ◦C. To measurethe metal concentrations, samples were digested with 6 N HCl and then analy-sis was performed using an atomic absorption (AA) spectrophotometer (PERKINELMER, AAnalyst 100) to measure the metal concentrations. The characteristicsof the contaminated soil were measured by the US EPA or ASTM methods (APHA,AWWA, and WPCF, 1995). They are listed in Table II.

2.3. EXPERIMENTAL SET-UP AND PROCEDURES

A series of experiments was conducted to investigate different parameters involvedin the biosurfactant foam technology in soil remediation. The experimental set-upis shown schematically in Figure 1. A plastic column (L = 25 cm, D = 2.5 cm) withmetal ends was equipped with circular porous stone plates, which enabled the foamgeneration in the presence of surfactant solution and air. A pump was used to feedthe surfactant solution. Two flow meters (I and II) were used to control the flow ofthe solution and the air before entering the foam generation column. The flow ratesof the surfactant solution and air could be varied independently in order to control


Figure 1. Schematic set-up of the column experiments.

the foam quality and generation rate. The foam exiting the column at atmosphericpressure was sampled for quality and stability analyses, after it reached a uniformand steady state in terms of bubble size and appearance. In each experiment, three50 mL samples of the generated foam were taken and left until all of the bubblescollapsed. The time required for collapsing half of the foam was recorded as anindicator of the foam stability, and the total liquid obtained from the collapsedfoam per total volume of the foam sample at atmospheric pressure was calculatedto indicate the foam quality. Foam quality was determined by the gas content of thefoam based on the volume ratio

Foam quality = gas volume

total foam volume× 100%.

All experiments were conducted at room temperature (25.0 ± 0.2 ◦C) and underatmospheric pressure. The effects of temperature and pressure on the foam volumewere negligible.

The soil column (L = 25 cm, D = 4 cm) was connected to the foam-generatingcolumn. The soil was packed uniformly into the column in five layers. The totalweight of soil was about 430 g (430 ± 5 g). A pressure gauge was installed upstreamof the soil column to measure the inlet pressure. A two-way valve was placed beforethe column to prevent injection of foam into soil before its uniformity and steady-state quality was reached. It was also used for taking foam samples before injectioninto the soil. The pressure in the column was recorded while changing the foamquality and the foam flow rate as well as the concentration of the biosurfactantsolution. The pressure gradient was then determined by dividing the pressure bythe length of the column. Results are shown for steady state.


The pore volume of the packed soil column, an important parameter for thefollowing experiments, was measured by saturating the soil column at a pressuregradient close to zero. The volume of water needed to saturate the soil column wasused to indicate the pore volume of the column.

Surfactant concentration is an important parameter influencing the heavy metalremoval efficiencies. It was found that metal removal efficiencies would increaselinearly with the increasing surfactant concentration below the critical micelle con-centration (CMC), and remain relatively constant above the CMC (Doong et al.,1998). Moreover, high concentrations of surfactant solutions (1–10%) often arerequired to overcome dilution and losses of the surfactant due to binding on thesoil (Oolman et al., 1995; Fountain et al., 1996). However, higher concentrationscan result in plugging by the dispersion of fine materials, or by the formation ofviscous emulsions (Nash, 1987; Rothmel et al., 1998). According to Chowdiah etal. (1998), from a physical perspective, to prevent problems due to channeling orsoil heaving, it is necessary to restrict injection pressure. On the basis of these con-siderations and also the results from the precedent experiments, 0.5% rhamnolipidsolution was chosen for the further experiments. Flow rates less than 5 mL/minwere used.

The flushing studies were conducted by varying the initial pH values (6.8, 8.0,and 10.0) of a 0.5% rhamnolipid solution. The pH value was adjusted by adding1 N NaOH solution. The original pH value of the rhamnolipid solution was around6.8. When the pH was lowered below this, the biosurfactant began to precipitateand therefore was not used in further experiments for foam generation. Foam witha controlled quality and flow rate was used to investigate the metal removal effi-ciencies. The foam exiting the soil column was collected for analysis of the metalconcentrations. Samples were collected according to the number of pore volumespassed through the contaminated soil and were left to collapse completely. Foamgenerated by Triton X-100 with the same concentration and pH value was used tocompare the metal removal efficiencies of the different surfactant foams. Surfactantsolutions without foam generation were also pumped through the column to eval-uate the metal removal efficiencies by aqueous surfactant solutions without foam.Distilled water with pH adjustment was used as the control to evaluate the flushingefficiencies without additives.

Columns flushed by the rhamnolipid foam were chosen to evaluate the massbalance after the experiments. At the end of the experiments, samples of soil weretaken from the flushed column to be analyzed for the heavy metal residual con-centrations. The results were compared with the results from effluent analyses toperform a mass balance on the metals.

Heavy metal concentrations were measured by atomic adsorption (AA) spec-trometry after acid (6 N and/or 12 N HCl) digestion. All the samples were analyzedtogether to ensure that the metal concentrations were measured under the sameconditions (such as same wavelength, standards, slope, etc.). The experiments weredone in triplicate and the metal concentrations in the effluents did not vary more


than 10%. The metal removal efficiency was determined on the basis of the initialmetal contents in the soil and all results are presented as percent metal removal.

3. Results and Discussion

3.1. FOAM PARAMETERS

Rhamnolipid solutions with different concentrations (0.5%, 1.0%, and 1.5%) withdifferent initial pH values (6.8, 8.0, and 10.0) were used to evaluate their foamabilityand the stability of the generated foam. High foam quality up to 99% was generatedand foam stability varied from 17 to 41 min. (Table III). The time it took for thefoam to collapse completely varied from 2 to 4 h., and even much longer.

As Rothmel et al. (1998) concluded, foam stability does not appear to be de-pendent upon inherent properties such as hydrophile-liphophile balance (HLB) andCMC. It was observed that foam quality and surfactant concentration had signifi-cant influences on the foam stability. Generally, foam stability increased with theincrease of surfactant concentration, and highest foam stability was produced withthe quality between 90 and 99%. Since the higher quality foam consists of largerbubbles with thinner liquid films, this effectively diminishes the capillary flow, pre-venting the rupture of the lamellar film between the adjacent bubbles. The foamweight is reduced with the increase of foam quality and the 99% quality foamoften has slugs of gas mixed in, which speeds the rupture of the bubbles, therebydecreasing the foam stability. From a theoretical perspective, the foam stabilitywill keep increasing with the increase of biosurfactant concentration until a thresh-old concentration is reached, then the stability will decrease with the increase ofthe biosurfactant concentration, due to the increasing weight. Obviously, 1.5% isnot such a threshold point. In general, these results showed that foams could be

TABLE III

Rhamnolipid foam stability and quality at 90, 95 and 99%

Foam Stability (minutes)Rhamnolipid Solutionconcentration pH values 90% 95% 99%

0.5% 6.8 17 24 20

8.0 18 24 17

10.0 18 25 21

1.0% 6.8 26 31 29

8.0 25 30 27

10.0 26 31 28

1.5% 6.8 31 39 32

8.0 32 39 33

10.0 32 41 35


generated from a low concentration of rhamnolipid solution (0.5% or perhaps less)with enough stability for injection.

3.2. PRESSURE GRADIENT BUILD-UP IN THE SOIL COLUMN

The experimental results (Figure 2) showed that the pressure gradient build-up in thesoil column varied from 0.3 to 6.6 kPa/cm for foam flushing through with different

Figure 2. Effect of initial foam quality and flow rate on pressure gradient (A: Rhamnolipid concen-tration = 0.5%; B: Rhamnolipid concentration = 1.0%; C: Rhamnolipid concentration = 1.5%).


qualities (from 95 to 99%), flow rates (from 10 to 30 mL/min) and rhamnolipid con-centrations (from 0.5 to 1.5%). It was observed that the pressure gradient increasedwith the increase of the flow rate and the decrease of foam quality. Higher qualityfoams exhibited lower pressure gradients during the flow because they consist oflarger bubbles with thinner films, which collapse more easily and rapidly whenpassing through the soil (Chowdiah et al., 1998). From the same curves, the effectsof foam flow rate on the pressure gradient build-up can be observed. The pressuregradient increased sharply with the increase of the foam flow rate. In Figure 3, anobservation is that the concentration of the surfactant solution, up to 1.5%, hadno significant effects, although theoretically, the pressure gradient will increasewith the surfactant solution concentration due to plugging of the column by thedispersion of fine materials, or by the formation of viscous emulsions.

These results showed that, in order to keep a lower pressure gradient in thesoil column, high quality foam from the biosurfactant solution with low concen-tration should be chosen for further experiments. At the same time, a low flowrate below 5 mL/min should be chosen to reduce the pressure gradient in the soilcolumn.

3.3. METAL REMOVAL EFFICIENCIES

In this phase, experiments were carried out to investigate the metal removal effi-ciencies by different extractants at different initial pH values (6.8, 8.0, and 10.0).Samples were collected separately according to the number of pore volumes ofsolution passed through the soil. Surfactant solutions at different pH values werefirst used to flush through the soil column without foam generation (the air line wasclosed). After passing 20 pore volumes, a 0.5% rhamnolipid solution with an initialpH of 10.0 showed a better result than with the other two initial pH values, whichremoved 61.7% of the Cd and 51.0% of the Ni (Figure 4a). Flushing experimentswere also conducted with another type of surfactant, Triton X-100, which is non-ionic (Figure 4b). There were only small differences in metal removal efficienciesfor Triton X-100 for the three tested pH values. Optimal removal was by 0.5%Triton X-100 solution (initial pH 10.0) which removed 52.8% of the Cd and 45.2%of the Ni after a 20-pore-volume flushing.

The possible mechanisms for the extraction of heavy metals by surfactantsinclude ion exchange, precipitation-dissolution and counterion association (Rosen,1979; Doong et al., 1998). It was postulated that the metals were removed byforming complexes with the surfactants on the soil surface, and detachment fromthe soil into the soil solution due to the lowering of the interfacial tension, andhence association with surfactant micelles. As expected, the anionic surfactantgave a better result because the cationic metals have an affinity for the negativelycharged surfactants, and also the biosurfactant could enable more metal removaldue to the more effective interfacial surface tension lowering. Solutions with highpH value seem to work better. Probably, this could be the result of a combination


Figure 3. Effect of rhamnolipid concentration on pressure gradient (A: Foam flow rate = 10 mL/min;B: Foam flow rate = 20 mL/min; C: Foam flow rate = 30 mL/min).

of the enhanced metal solubility and surfactant activity with the increase of thesolution pH value. Increased pH favors the ionization of the carboxyl portion ofthe rhamnolipid (RCOO−) in the solution. This enhances water solubility, providesmore binding sites for the metal cations on the biosurfactant, and leads to increasedmetal solubility. Moreover, it has been shown that interfacial tension (IFT) can befurther reduced by increasing the solution pH (Lin et al., 1987).

A series of experiments was performed using foam generated by 0.5% rham-nolipid solutions at different pH values (6.8, 8.0, and 10.0) to investigate the


(a)

(b)

Figure 4. Metal removal efficiencies by 0.5% surfactant solutions at different initial pH values(a: Rhamnolipid solution, b: Triton X-100 solution).

metal removal efficiencies. The best removal rate, 73.2% of the Cd and 68.1%of the Ni, was achieved by adjusting the initial solution pH value to 10.0, af-ter a 20-pore-volume flushing. A 0.5% solution of Triton X-100 adjusted to pH10.0 was used to generate foam and was injected into the column. Using these


Figure 5. Metal removal efficiencies by rhamnolipid foam (Initial pH = 10.0).

conditions, 65.5% of the Cd and 57.3% of the Ni were removed after a 20-pore-volume flushing. Figure 5 illustrates the trend of metal removal efficiency withfoam flushing through the column. It can be observed that the metal removal effi-ciency increased sharply within the first 5 pore volumes. Then it remained relativelyconstant when more foam passed through. Most of the metals removed were con-centrated in the first 5 pore volumes. This indicates that in the field applications, thismethod may be more cost efficient because of the time saved and the low usage ofsurfactants.

By comparing the metal removal efficiencies with surfactant solutions andfoams, it is observed that over 11% more of the Cd and 17% more of the Niwere removed by rhamnolipid foam than rhamnolipid solution with the same con-centration (Figure 6), while over 12% more of the Cd and 12% more of the Ni wereremoved by Triton X-100 foam than Triton X-100 solution (Figure 7). Transferfrom aqueous solution to foam greatly enhanced the metal removal efficiencies bythe same concentration of both tested surfactants.

These enhancements can be explained from several aspects. As known, the metalremoval efficiencies of soil flushing depend, to a large extent, on the contact timeand contact surface area of the surfactant with contaminated soil. The use of foamproduces a more uniform and efficient contact of the surfactant with the metals, andrestrains the possible channeling in the column effectively, which results in highermetal removal efficiencies. First, the injection of surfactant and air at the same timehas a significant effect on the mobility of fluid flowing in a porous medium, becauseof the scattering nature of foam. It was demonstrated that the effective permeability


Figure 6. Metal removal efficiencies by 0.5% rhamnolipid solution and foam (A: Cd removalefficiencies; B: Ni removal efficiencies).

Figure 7. Metal removal efficiencies by 0.5% Triton X-100 solution and foam.

of the porous medium to each phase is significantly decreased when foam is present(Bernard and Holm, 1964; Bernard et al., 1965). The reduction of fluid mobilityincreases fluid residence time and the contact efficiency between the surfactant andthe contaminated soil. Secondly, when transferred from liquid to foam, the fluiddensity is reduced, for example, when a 95% quality foam is used, the fluid densityis reduced to around 0.05 times of the liquid, so it can easily overcome gravitationaleffects and flow through most of the regions in the column. Thus the channelingis inhibited and the homogeneity of the flood is improved. Another advantage ofusing foam is the large surface area of the liquid in the foam compared with theair-liquid interface without foam. For example, when the surfactant solution is usedto produce a foam with a quality of 95%, the volume is increased by 20 times. Thedirect result of which is the increased surface area of about 7 times (on the basisof the assumption that the foam bubbles are of spherical shape). It leads to a moreefficient contact of surfactant with the metals. Moreover, the soil wettability anddesorption can potentially be improved by the surfactants used to generate the foam(Chowdiah et al., 1998).


Figure 8. Metal removal efficiencies by different extractants (Initial pH = 10.0).

To determine the metal removal efficiencies without additives, control experi-ments were carried out using distilled water with a correspondently adjusted pHvalue of 10.0 (Figure 8). It removed 17.8% of the Cd and 18.7% of the Ni (Figure 8).It worked mainly as a physical dissolvent and removed the metals that precipitated,but were not bound strongly on the surface of the soil particles.

3.4. HEAVY METAL MASS BALANCES

Soil samples were taken from each of the chosen columns flushed by rhamnolipidfoam after the experiments. AA analysis was performed to test the residual metalconcentrations in the soil. The results showed that the mass balances were goodduring the experiments. On average, 1.6% of the Cd and 0.4% of the Ni were notaccounted for (Table IV). This may be caused by the analytical and/or operationalerrors, which are unavoidable but expected during the experiments.

TABLE IV

Cadmium and nickel mass balances

Remaining UnaccountedMetal Removed in soil Total for

Cd 73.2% 25.2% 98.4% 1.6%

Ni 68.1% 31.5% 99.6% 0.4%


4. Conclusions

This study demonstrates that rhamnolipid foam technology can be used successfullyin the remediation of heavy metal contaminated sandy soil. Although this study wasinvestigated in idealized laboratory conditions, the results from this study illustratesome general considerations that are important for the use of foams for the soilflushing during the in field applications. High quality foam with enough stabilitycan be generated by the rhamnolipid solution with a low concentration, and thepressure gradient build-up in soil caused by injection of foam could be restrictedto the ideal value scope. Promising results showed that foam technology is moreeffective in metal removal than conventional soil flushing by surfactant solutions.

However, it is not for all-purpose operations. For example, it would not workwell in contaminated soils with low porosity such as clays. Also, test results infield applications tend to vary more from those obtained in the lab scale benchtests, because the capacity of foam to extract and mobilize the metals from soilsis a function of various parameters, such as the pH, soil matrix characteristics, themetal speciation and residence time as well as the surfactant type and concentration.Extensive future experiments in which both the soil and the foam are involved haveto be done. Also further research is needed to evaluate the cost effectiveness of thistechnology in field applications.

Acknowledgements

The authors would like to acknowledge the financial support of NSERC and theConcordia University Faculty of Engineering and Computer Science Faculty Sup-port. The authors would also like to thank Jeneil Biosurfactant Co. for the supplyof the JBR biosurfactant.

References

American Public Health Association (APHA), American Water Works Association (AWWA) andWater Pollution Control Federation (WPCF): 1995, ‘Standard Methods for the Examination ofWater and Wastewater’, 17th edn., APHA, Washington, DC.

Bernard, G. G. and Holm, L. W.: 1964, ‘Effect of foam on permeability of porous media to gas’, Soc.Pet. Eng. J. 19, 267–274.

Bernard, G. G., Holm, L. W. and Jacobs, W. L.: 1965, ‘Effect of foam on trapped gas saturation andon permeability of porous media to water’, Soc. Pet. Eng. J. 5, 295–300.

Chowdiah, P., Misra, B. R., Kilbane, J. J.,II, Srivastava, V. J. and Hayes, T. D.: 1998, ‘Foam propagationthrough soils for enhanced in-situ remediation’, J. Hazard. Mat. 62, 265–280.

Doong, R. A., Wu, Y. W. and Lei, W. G.: 1998, ‘Surfactant enhanced remediation of cadmiumcontaminated soils’, Wat. Sci. Technol. 37(8), 65–71.

Evanko, C. R. and Dzombak, A. D.: 1997, ‘Remediation of Metals-Contaminated Soils and Ground-water’, TE-97-01, Ground-Water Remediation Technologies Analysis Center.


Fountain, J. C., Starr, R. S., Middleton, T., Beikirch, M., Taylor, C. and Hodge, D. S.: 1996, ‘Acontrolled field test of surfactant enhanced aquifer remediation’, Ground Wat. 34(5), 910–916.

Gillman, G. P. and Sumpter, E. A.: 1986, ‘Modification to the compulsive exchange method formeasuring exchange characteristics of soils’, Aust. J. Soil Res. 24, 61–66.

JENEIL Biosurfactant Co., LLC.: 2001, ‘Material safety data sheet for JBR425’, Retrived fromhttp://www.biosurfactant.com/downloads/jbr425msds.pdf.

Jeong, S. W., Corapcioglu, M. Y. and Roosevelt, S. E.: 2000, ‘Micromodel study of surfactant foamremediation of residual trichlorothylene’, Environ. Sci. Technol. 34, 3456–3461.

Kilbane, J. J., II, Chowdiah, P., Kayser, K. J., Misra, B., Jackowski, K. A., Srivastava, V. J., Sethu,G. N., Nikolov, A. D., Wasan, D. T. and Hayes, T. D.: 1997, ‘Remediation of contaminated soilsusing foams’, Land Contamin. Reclam. 5, 41–54.

Lin, F. F. J., Besserer, G. L. and Pitts, M. J.: 1987, ‘Laboratory evaluation of crosslinked polymer andalkailine-polymer-surfactant flood’, J. Can. Petrol. Technol. 26, 54–65.

Miller, R. M.: 1995, ‘Biosurfactant-facilitated remediation of metal-contaminated soils’, Environ.Health Perspect 103(suppl 1), 59–61.

Mulligan, C. N., Yong, R. N. and Gibbs, B. F.: 1999, ‘On the use of biosurfactants for the removal ofheavy metals from oil-contaminated soil’, Environ. Prog. 18(1), 50–54.

Mulligan, C. N, Yong, R. N. and Gibbs, B. F.: 2001a, ‘Heavy metal removal from sediments bybiosurfactants’, J. Hazard. Mat. 85, 111–125.

Mulligan, C. N., Yong, R. N. and Gibbs, B. F.: 2001b, ‘An evaluation of technologies for the heavymetal remediation of dredged sediments’, J. Hazard. Mat. 85, 145–163.

Mulligan, C. N. and Eftekhari, F.: 2003, ‘Remediation with surfactant foam of pcp-contaminated soil’,Eng. Geol. 70, 269–279.

Nash, J. H.: 1987, Field Studies of In Situ Soil Washing, Environmental Protection Agency, Cincinnati,Ohio, Hazardous Waste Engineering Research Laboratory. Report Number PA/600/2-87/110,PB88-146808.

Oolman, T., Godard, S. T., Pope, G. A., Jin, M. and Kirchner, K.: 1995, ‘DNAPL flow behavior in acontaminated aquifer: evaluation of field data’, Ground Wat. Monit. Remed. 15(4), 125–137.

Peters, R. W., Enzien, W. V., Bouillard, J. X., Frank, J. R., Srivastava, V. J., Kilbane II, J. J. and Hayes,T. D.: 1994, ‘Nonaqueous-phase-liquids-contaminated soil/groundwater remediation using foamsin the in-situ remediation’, in G.W. Gee and N.R Wing (eds), Scientific Basis for Current andFuture Ttechnologies, Battelle Press, Columbus, OH, pp. 1067–1087.

Rosen, M. J.: 1979, Surfactants and Interfacial Phenomena, Wiley, New York.Rothmel, R. K., Peters, R. W., Martin, E. St. and Deflaun, M. F.: 1998, ‘Surfactant foam/bioaugmen-

tation technology for in situ treatment of TCE-DNAPLs’, Environ. Sci. Technol. 32, 1667–1675.Roundhill, D. M.: 2001, ‘Extraction of Metals from Soil and Waters’, John P.F, Jr. (ed.), Modern

Inorganic Chemistry, Kluwer Academic/Plenum Publishers, Boston, MA.Selim, H. M. and Michael, C. A.: 1996, Reactivity and Transport of Heavy Metals in Soils, Lewis

Publishers, Boca Raton, Florida, U.S.A.Sigma: 1993, ‘Sigma Product Information Sheet: Triton X-100’, Retrived from http://www .sig-

maaldrich.com/sigma/prodata/t6878.htm.Torrens, J. L., Herman, D. C. and Miller-Maier, R. M.: 1998, ‘Biosurfactant (rhamnolipid) sorption

and the impact on rhamnolipid-facilitated removal of cadmium from various soils under saturatedflow conditions’, Environ. Sci. Technol. 32(6), 776–781.

Tsujii, K.: 1998, Surface Activity, Principles, Phenomena, and Applications. Academic Press, Boston,MA.

Vignon, B. W. and Rubin, A. J.: 1989, ‘Practical considerations in the surfactant-aided mobilizationof contaminants in aquifers’, J. Water Pollut. Control 61, 1233–1240.

rhamnolipid foam enhanced remediation of cadmium

Documents

metal removal efciencies

heavy metal removal

metal removal efciency

initial ph values

pressure gradient build

foam enhanced remediation

foam ow rate

high quality foam